Perplexity
- Category
- Verification
- Type
- commercial
- Scope
- citation-backed answering and source attribution only
- Source
- Vendor site
- Last reviewed
- 2026-10-17 (what this means)
Citation check incomplete — recheck later.
Scope note: the Comet browser and its agent are covered in a separate entry, Perplexity Comet.
Summary
Perplexity is a commercial AI answer engine; this entry covers only its citation and source-attribution behavior. In independent evaluations (CJR Tow Center, 2025; Findings of EMNLP 2023; DeepResearch Bench, ICLR 2026; DeepResearchGym, ICTIR 2026; Deep Research Comparator, WWW 2026 Companion; Mind2Web 2, NeurIPS 2025), results varied by product mode: its Deep Research mode scored comparatively high on citation accuracy, while news-attribution tests found a share of incorrect or syndicated citations. Several publishers have filed copyright lawsuits against Perplexity that are pending; these are allegations, not findings. Informational only; not a guarantee, endorsement, or certification.
Overall score: 54 · Evidence grade: B · Independent sources: 6 performance, 3 other (court records)
Factor breakdown
| Factor | Weight | Score | Note |
|---|---|---|---|
| Completion rate | 30% | In CJR's news-attribution test, 37% of answers were incorrect (the lowest error rate of 8 tools tested); Mind2Web 2 success rates were 8% (Pro Search) and 15% (Deep Research). Judgment call. | |
| Disputes | 20% | Several publishers have filed copyright lawsuits that are pending (allegations, not findings); CJR also reported attribution to syndicated copies. Judgment call. | |
| Adherence to limits | 25% | CJR's results suggest, in CJR's words, that Perplexity 'may have disregarded' one publisher's robots.txt preferences; Perplexity states it respects robots.txt. Vendor plan-limit descriptions appeared inconsistent. Judgment call. | |
| Evidence quality | 25% | Six independent performance sources (five in peer-reviewed venues, plus CJR), covering different product modes and dates (2023–2025). Judgment call. |
Overall = 0.30×55 + 0.20×35 + 0.25×45 + 0.25×75 = 53.5, shown as 54 (exact .5 rounds up). See the methodology.
Legal disputes (filed; pending)
The following are allegations in pending lawsuits, recorded from court records. They are not findings.
- Dow Jones & Co. and NYP Holdings filed suit (S.D.N.Y. 1:24-cv-07984), alleging copyright infringement, false designation of origin and trademark dilution. The court denied Perplexity's motion to dismiss or transfer on 2025-08-21; the case is pending.
- Encyclopaedia Britannica and Merriam-Webster filed suit on 2025-09-10 (S.D.N.Y. 1:25-cv-07546), alleging copyright infringement; the case is pending.
- Chicago Tribune Company filed suit on 2025-12-04, and The New York Times Company filed suit on 2025-12-05 (S.D.N.Y. 1:25-cv-10106), alleging copyright infringement; both are pending.
Uncertainties
- Results differ by product mode (standard, Pro, Deep Research, API) and test date (2023–2025).
- Vendor help-center pages could not be fetched directly during review; plan limits are not verified.
- Lawsuits are pending allegations, not findings.
- The robots.txt point is CJR's interpretation, stated by CJR as a suggestion.
- The overall score is exactly 53.5; exact .5 rounds up (Decision 0010).
- Three peer-reviewed health-information studies (PLoS ONE 2025, PeerJ 2025, PLoS ONE 2026; the first two share authors) rated Perplexity highly on readability and quality scales, including a June 2026 test of the free Sonar model; not counted because they check that sources are listed, not that sources support the answer.
- Citation check incomplete — recheck later (0056: five 2025–2026 candidates read, none in scope; Semantic Scholar still unavailable).
Sources
- CJR Tow Center (2025): news citation/attribution test; free Perplexity and Perplexity Pro — independent performance, accessed 2026-10-16
- Liu et al., Findings of EMNLP 2023: citation recall/precision (perplexity.ai, 2023) — independent performance, accessed 2026-10-16
- DeepResearch Bench (ICLR 2026): Perplexity Deep Research citation accuracy — independent performance, accessed 2026-10-16
- DeepResearchGym (ICTIR 2026): sonar-deepresearch citation precision/recall — independent performance, accessed 2026-10-16
- Deep Research Comparator (WWW 2026 Companion): human span annotations — independent performance, accessed 2026-10-16
- Mind2Web 2 (NeurIPS 2025 D&B): task success with source attribution — independent performance, accessed 2026-10-16
- Court record: Dow Jones & Co. and NYP Holdings v. Perplexity AI (S.D.N.Y.), opinion and order — independent other, accessed 2026-10-16
- Docket: Encyclopaedia Britannica and Merriam-Webster v. Perplexity AI (S.D.N.Y.) — independent other, accessed 2026-10-16
- Court record: The New York Times Company related-case statement (S.D.N.Y.) — independent other, accessed 2026-10-16
- Perplexity Help Center: Pro Search and citations — vendor or maintainer, accessed 2026-10-16
- Perplexity Help Center: plan limits — vendor or maintainer, accessed 2026-10-16